Loading...
1 Linux kernel release 3.x <http://kernel.org/>
2
3These are the release notes for Linux version 3. Read them carefully,
4as they tell you what this is all about, explain how to install the
5kernel, and what to do if something goes wrong.
6
7WHAT IS LINUX?
8
9 Linux is a clone of the operating system Unix, written from scratch by
10 Linus Torvalds with assistance from a loosely-knit team of hackers across
11 the Net. It aims towards POSIX and Single UNIX Specification compliance.
12
13 It has all the features you would expect in a modern fully-fledged Unix,
14 including true multitasking, virtual memory, shared libraries, demand
15 loading, shared copy-on-write executables, proper memory management,
16 and multistack networking including IPv4 and IPv6.
17
18 It is distributed under the GNU General Public License - see the
19 accompanying COPYING file for more details.
20
21ON WHAT HARDWARE DOES IT RUN?
22
23 Although originally developed first for 32-bit x86-based PCs (386 or higher),
24 today Linux also runs on (at least) the Compaq Alpha AXP, Sun SPARC and
25 UltraSPARC, Motorola 68000, PowerPC, PowerPC64, ARM, Hitachi SuperH, Cell,
26 IBM S/390, MIPS, HP PA-RISC, Intel IA-64, DEC VAX, AMD x86-64, AXIS CRIS,
27 Xtensa, Tilera TILE, AVR32 and Renesas M32R architectures.
28
29 Linux is easily portable to most general-purpose 32- or 64-bit architectures
30 as long as they have a paged memory management unit (PMMU) and a port of the
31 GNU C compiler (gcc) (part of The GNU Compiler Collection, GCC). Linux has
32 also been ported to a number of architectures without a PMMU, although
33 functionality is then obviously somewhat limited.
34 Linux has also been ported to itself. You can now run the kernel as a
35 userspace application - this is called UserMode Linux (UML).
36
37DOCUMENTATION:
38
39 - There is a lot of documentation available both in electronic form on
40 the Internet and in books, both Linux-specific and pertaining to
41 general UNIX questions. I'd recommend looking into the documentation
42 subdirectories on any Linux FTP site for the LDP (Linux Documentation
43 Project) books. This README is not meant to be documentation on the
44 system: there are much better sources available.
45
46 - There are various README files in the Documentation/ subdirectory:
47 these typically contain kernel-specific installation notes for some
48 drivers for example. See Documentation/00-INDEX for a list of what
49 is contained in each file. Please read the Changes file, as it
50 contains information about the problems, which may result by upgrading
51 your kernel.
52
53 - The Documentation/DocBook/ subdirectory contains several guides for
54 kernel developers and users. These guides can be rendered in a
55 number of formats: PostScript (.ps), PDF, HTML, & man-pages, among others.
56 After installation, "make psdocs", "make pdfdocs", "make htmldocs",
57 or "make mandocs" will render the documentation in the requested format.
58
59INSTALLING the kernel source:
60
61 - If you install the full sources, put the kernel tarball in a
62 directory where you have permissions (eg. your home directory) and
63 unpack it:
64
65 gzip -cd linux-3.X.tar.gz | tar xvf -
66
67 or
68
69 bzip2 -dc linux-3.X.tar.bz2 | tar xvf -
70
71 Replace "X" with the version number of the latest kernel.
72
73 Do NOT use the /usr/src/linux area! This area has a (usually
74 incomplete) set of kernel headers that are used by the library header
75 files. They should match the library, and not get messed up by
76 whatever the kernel-du-jour happens to be.
77
78 - You can also upgrade between 3.x releases by patching. Patches are
79 distributed in the traditional gzip and the newer bzip2 format. To
80 install by patching, get all the newer patch files, enter the
81 top level directory of the kernel source (linux-3.X) and execute:
82
83 gzip -cd ../patch-3.x.gz | patch -p1
84
85 or
86
87 bzip2 -dc ../patch-3.x.bz2 | patch -p1
88
89 Replace "x" for all versions bigger than the version "X" of your current
90 source tree, _in_order_, and you should be ok. You may want to remove
91 the backup files (some-file-name~ or some-file-name.orig), and make sure
92 that there are no failed patches (some-file-name# or some-file-name.rej).
93 If there are, either you or I have made a mistake.
94
95 Unlike patches for the 3.x kernels, patches for the 3.x.y kernels
96 (also known as the -stable kernels) are not incremental but instead apply
97 directly to the base 3.x kernel. For example, if your base kernel is 3.0
98 and you want to apply the 3.0.3 patch, you must not first apply the 3.0.1
99 and 3.0.2 patches. Similarly, if you are running kernel version 3.0.2 and
100 want to jump to 3.0.3, you must first reverse the 3.0.2 patch (that is,
101 patch -R) _before_ applying the 3.0.3 patch. You can read more on this in
102 Documentation/applying-patches.txt
103
104 Alternatively, the script patch-kernel can be used to automate this
105 process. It determines the current kernel version and applies any
106 patches found.
107
108 linux/scripts/patch-kernel linux
109
110 The first argument in the command above is the location of the
111 kernel source. Patches are applied from the current directory, but
112 an alternative directory can be specified as the second argument.
113
114 - Make sure you have no stale .o files and dependencies lying around:
115
116 cd linux
117 make mrproper
118
119 You should now have the sources correctly installed.
120
121SOFTWARE REQUIREMENTS
122
123 Compiling and running the 3.x kernels requires up-to-date
124 versions of various software packages. Consult
125 Documentation/Changes for the minimum version numbers required
126 and how to get updates for these packages. Beware that using
127 excessively old versions of these packages can cause indirect
128 errors that are very difficult to track down, so don't assume that
129 you can just update packages when obvious problems arise during
130 build or operation.
131
132BUILD directory for the kernel:
133
134 When compiling the kernel, all output files will per default be
135 stored together with the kernel source code.
136 Using the option "make O=output/dir" allow you to specify an alternate
137 place for the output files (including .config).
138 Example:
139
140 kernel source code: /usr/src/linux-3.X
141 build directory: /home/name/build/kernel
142
143 To configure and build the kernel, use:
144
145 cd /usr/src/linux-3.X
146 make O=/home/name/build/kernel menuconfig
147 make O=/home/name/build/kernel
148 sudo make O=/home/name/build/kernel modules_install install
149
150 Please note: If the 'O=output/dir' option is used, then it must be
151 used for all invocations of make.
152
153CONFIGURING the kernel:
154
155 Do not skip this step even if you are only upgrading one minor
156 version. New configuration options are added in each release, and
157 odd problems will turn up if the configuration files are not set up
158 as expected. If you want to carry your existing configuration to a
159 new version with minimal work, use "make oldconfig", which will
160 only ask you for the answers to new questions.
161
162 - Alternative configuration commands are:
163
164 "make config" Plain text interface.
165
166 "make menuconfig" Text based color menus, radiolists & dialogs.
167
168 "make nconfig" Enhanced text based color menus.
169
170 "make xconfig" X windows (Qt) based configuration tool.
171
172 "make gconfig" X windows (Gtk) based configuration tool.
173
174 "make oldconfig" Default all questions based on the contents of
175 your existing ./.config file and asking about
176 new config symbols.
177
178 "make silentoldconfig"
179 Like above, but avoids cluttering the screen
180 with questions already answered.
181 Additionally updates the dependencies.
182
183 "make defconfig" Create a ./.config file by using the default
184 symbol values from either arch/$ARCH/defconfig
185 or arch/$ARCH/configs/${PLATFORM}_defconfig,
186 depending on the architecture.
187
188 "make ${PLATFORM}_defconfig"
189 Create a ./.config file by using the default
190 symbol values from
191 arch/$ARCH/configs/${PLATFORM}_defconfig.
192 Use "make help" to get a list of all available
193 platforms of your architecture.
194
195 "make allyesconfig"
196 Create a ./.config file by setting symbol
197 values to 'y' as much as possible.
198
199 "make allmodconfig"
200 Create a ./.config file by setting symbol
201 values to 'm' as much as possible.
202
203 "make allnoconfig" Create a ./.config file by setting symbol
204 values to 'n' as much as possible.
205
206 "make randconfig" Create a ./.config file by setting symbol
207 values to random values.
208
209 You can find more information on using the Linux kernel config tools
210 in Documentation/kbuild/kconfig.txt.
211
212 - NOTES on "make config":
213
214 - Having unnecessary drivers will make the kernel bigger, and can
215 under some circumstances lead to problems: probing for a
216 nonexistent controller card may confuse your other controllers
217
218 - Compiling the kernel with "Processor type" set higher than 386
219 will result in a kernel that does NOT work on a 386. The
220 kernel will detect this on bootup, and give up.
221
222 - A kernel with math-emulation compiled in will still use the
223 coprocessor if one is present: the math emulation will just
224 never get used in that case. The kernel will be slightly larger,
225 but will work on different machines regardless of whether they
226 have a math coprocessor or not.
227
228 - The "kernel hacking" configuration details usually result in a
229 bigger or slower kernel (or both), and can even make the kernel
230 less stable by configuring some routines to actively try to
231 break bad code to find kernel problems (kmalloc()). Thus you
232 should probably answer 'n' to the questions for "development",
233 "experimental", or "debugging" features.
234
235COMPILING the kernel:
236
237 - Make sure you have at least gcc 3.2 available.
238 For more information, refer to Documentation/Changes.
239
240 Please note that you can still run a.out user programs with this kernel.
241
242 - Do a "make" to create a compressed kernel image. It is also
243 possible to do "make install" if you have lilo installed to suit the
244 kernel makefiles, but you may want to check your particular lilo setup first.
245
246 To do the actual install, you have to be root, but none of the normal
247 build should require that. Don't take the name of root in vain.
248
249 - If you configured any of the parts of the kernel as `modules', you
250 will also have to do "make modules_install".
251
252 - Verbose kernel compile/build output:
253
254 Normally, the kernel build system runs in a fairly quiet mode (but not
255 totally silent). However, sometimes you or other kernel developers need
256 to see compile, link, or other commands exactly as they are executed.
257 For this, use "verbose" build mode. This is done by inserting
258 "V=1" in the "make" command. E.g.:
259
260 make V=1 all
261
262 To have the build system also tell the reason for the rebuild of each
263 target, use "V=2". The default is "V=0".
264
265 - Keep a backup kernel handy in case something goes wrong. This is
266 especially true for the development releases, since each new release
267 contains new code which has not been debugged. Make sure you keep a
268 backup of the modules corresponding to that kernel, as well. If you
269 are installing a new kernel with the same version number as your
270 working kernel, make a backup of your modules directory before you
271 do a "make modules_install".
272
273 Alternatively, before compiling, use the kernel config option
274 "LOCALVERSION" to append a unique suffix to the regular kernel version.
275 LOCALVERSION can be set in the "General Setup" menu.
276
277 - In order to boot your new kernel, you'll need to copy the kernel
278 image (e.g. .../linux/arch/i386/boot/bzImage after compilation)
279 to the place where your regular bootable kernel is found.
280
281 - Booting a kernel directly from a floppy without the assistance of a
282 bootloader such as LILO, is no longer supported.
283
284 If you boot Linux from the hard drive, chances are you use LILO, which
285 uses the kernel image as specified in the file /etc/lilo.conf. The
286 kernel image file is usually /vmlinuz, /boot/vmlinuz, /bzImage or
287 /boot/bzImage. To use the new kernel, save a copy of the old image
288 and copy the new image over the old one. Then, you MUST RERUN LILO
289 to update the loading map!! If you don't, you won't be able to boot
290 the new kernel image.
291
292 Reinstalling LILO is usually a matter of running /sbin/lilo.
293 You may wish to edit /etc/lilo.conf to specify an entry for your
294 old kernel image (say, /vmlinux.old) in case the new one does not
295 work. See the LILO docs for more information.
296
297 After reinstalling LILO, you should be all set. Shutdown the system,
298 reboot, and enjoy!
299
300 If you ever need to change the default root device, video mode,
301 ramdisk size, etc. in the kernel image, use the 'rdev' program (or
302 alternatively the LILO boot options when appropriate). No need to
303 recompile the kernel to change these parameters.
304
305 - Reboot with the new kernel and enjoy.
306
307IF SOMETHING GOES WRONG:
308
309 - If you have problems that seem to be due to kernel bugs, please check
310 the file MAINTAINERS to see if there is a particular person associated
311 with the part of the kernel that you are having trouble with. If there
312 isn't anyone listed there, then the second best thing is to mail
313 them to me (torvalds@linux-foundation.org), and possibly to any other
314 relevant mailing-list or to the newsgroup.
315
316 - In all bug-reports, *please* tell what kernel you are talking about,
317 how to duplicate the problem, and what your setup is (use your common
318 sense). If the problem is new, tell me so, and if the problem is
319 old, please try to tell me when you first noticed it.
320
321 - If the bug results in a message like
322
323 unable to handle kernel paging request at address C0000010
324 Oops: 0002
325 EIP: 0010:XXXXXXXX
326 eax: xxxxxxxx ebx: xxxxxxxx ecx: xxxxxxxx edx: xxxxxxxx
327 esi: xxxxxxxx edi: xxxxxxxx ebp: xxxxxxxx
328 ds: xxxx es: xxxx fs: xxxx gs: xxxx
329 Pid: xx, process nr: xx
330 xx xx xx xx xx xx xx xx xx xx
331
332 or similar kernel debugging information on your screen or in your
333 system log, please duplicate it *exactly*. The dump may look
334 incomprehensible to you, but it does contain information that may
335 help debugging the problem. The text above the dump is also
336 important: it tells something about why the kernel dumped code (in
337 the above example, it's due to a bad kernel pointer). More information
338 on making sense of the dump is in Documentation/oops-tracing.txt
339
340 - If you compiled the kernel with CONFIG_KALLSYMS you can send the dump
341 as is, otherwise you will have to use the "ksymoops" program to make
342 sense of the dump (but compiling with CONFIG_KALLSYMS is usually preferred).
343 This utility can be downloaded from
344 ftp://ftp.<country>.kernel.org/pub/linux/utils/kernel/ksymoops/ .
345 Alternatively, you can do the dump lookup by hand:
346
347 - In debugging dumps like the above, it helps enormously if you can
348 look up what the EIP value means. The hex value as such doesn't help
349 me or anybody else very much: it will depend on your particular
350 kernel setup. What you should do is take the hex value from the EIP
351 line (ignore the "0010:"), and look it up in the kernel namelist to
352 see which kernel function contains the offending address.
353
354 To find out the kernel function name, you'll need to find the system
355 binary associated with the kernel that exhibited the symptom. This is
356 the file 'linux/vmlinux'. To extract the namelist and match it against
357 the EIP from the kernel crash, do:
358
359 nm vmlinux | sort | less
360
361 This will give you a list of kernel addresses sorted in ascending
362 order, from which it is simple to find the function that contains the
363 offending address. Note that the address given by the kernel
364 debugging messages will not necessarily match exactly with the
365 function addresses (in fact, that is very unlikely), so you can't
366 just 'grep' the list: the list will, however, give you the starting
367 point of each kernel function, so by looking for the function that
368 has a starting address lower than the one you are searching for but
369 is followed by a function with a higher address you will find the one
370 you want. In fact, it may be a good idea to include a bit of
371 "context" in your problem report, giving a few lines around the
372 interesting one.
373
374 If you for some reason cannot do the above (you have a pre-compiled
375 kernel image or similar), telling me as much about your setup as
376 possible will help. Please read the REPORTING-BUGS document for details.
377
378 - Alternatively, you can use gdb on a running kernel. (read-only; i.e. you
379 cannot change values or set break points.) To do this, first compile the
380 kernel with -g; edit arch/i386/Makefile appropriately, then do a "make
381 clean". You'll also need to enable CONFIG_PROC_FS (via "make config").
382
383 After you've rebooted with the new kernel, do "gdb vmlinux /proc/kcore".
384 You can now use all the usual gdb commands. The command to look up the
385 point where your system crashed is "l *0xXXXXXXXX". (Replace the XXXes
386 with the EIP value.)
387
388 gdb'ing a non-running kernel currently fails because gdb (wrongly)
389 disregards the starting offset for which the kernel is compiled.
390
1 Linux kernel release 4.x <http://kernel.org/>
2
3These are the release notes for Linux version 4. Read them carefully,
4as they tell you what this is all about, explain how to install the
5kernel, and what to do if something goes wrong.
6
7WHAT IS LINUX?
8
9 Linux is a clone of the operating system Unix, written from scratch by
10 Linus Torvalds with assistance from a loosely-knit team of hackers across
11 the Net. It aims towards POSIX and Single UNIX Specification compliance.
12
13 It has all the features you would expect in a modern fully-fledged Unix,
14 including true multitasking, virtual memory, shared libraries, demand
15 loading, shared copy-on-write executables, proper memory management,
16 and multistack networking including IPv4 and IPv6.
17
18 It is distributed under the GNU General Public License - see the
19 accompanying COPYING file for more details.
20
21ON WHAT HARDWARE DOES IT RUN?
22
23 Although originally developed first for 32-bit x86-based PCs (386 or higher),
24 today Linux also runs on (at least) the Compaq Alpha AXP, Sun SPARC and
25 UltraSPARC, Motorola 68000, PowerPC, PowerPC64, ARM, Hitachi SuperH, Cell,
26 IBM S/390, MIPS, HP PA-RISC, Intel IA-64, DEC VAX, AMD x86-64, AXIS CRIS,
27 Xtensa, Tilera TILE, AVR32, ARC and Renesas M32R architectures.
28
29 Linux is easily portable to most general-purpose 32- or 64-bit architectures
30 as long as they have a paged memory management unit (PMMU) and a port of the
31 GNU C compiler (gcc) (part of The GNU Compiler Collection, GCC). Linux has
32 also been ported to a number of architectures without a PMMU, although
33 functionality is then obviously somewhat limited.
34 Linux has also been ported to itself. You can now run the kernel as a
35 userspace application - this is called UserMode Linux (UML).
36
37DOCUMENTATION:
38
39 - There is a lot of documentation available both in electronic form on
40 the Internet and in books, both Linux-specific and pertaining to
41 general UNIX questions. I'd recommend looking into the documentation
42 subdirectories on any Linux FTP site for the LDP (Linux Documentation
43 Project) books. This README is not meant to be documentation on the
44 system: there are much better sources available.
45
46 - There are various README files in the Documentation/ subdirectory:
47 these typically contain kernel-specific installation notes for some
48 drivers for example. See Documentation/00-INDEX for a list of what
49 is contained in each file. Please read the Changes file, as it
50 contains information about the problems, which may result by upgrading
51 your kernel.
52
53 - The Documentation/DocBook/ subdirectory contains several guides for
54 kernel developers and users. These guides can be rendered in a
55 number of formats: PostScript (.ps), PDF, HTML, & man-pages, among others.
56 After installation, "make psdocs", "make pdfdocs", "make htmldocs",
57 or "make mandocs" will render the documentation in the requested format.
58
59INSTALLING the kernel source:
60
61 - If you install the full sources, put the kernel tarball in a
62 directory where you have permissions (e.g. your home directory) and
63 unpack it:
64
65 xz -cd linux-4.X.tar.xz | tar xvf -
66
67 Replace "X" with the version number of the latest kernel.
68
69 Do NOT use the /usr/src/linux area! This area has a (usually
70 incomplete) set of kernel headers that are used by the library header
71 files. They should match the library, and not get messed up by
72 whatever the kernel-du-jour happens to be.
73
74 - You can also upgrade between 4.x releases by patching. Patches are
75 distributed in the xz format. To install by patching, get all the
76 newer patch files, enter the top level directory of the kernel source
77 (linux-4.X) and execute:
78
79 xz -cd ../patch-4.x.xz | patch -p1
80
81 Replace "x" for all versions bigger than the version "X" of your current
82 source tree, _in_order_, and you should be ok. You may want to remove
83 the backup files (some-file-name~ or some-file-name.orig), and make sure
84 that there are no failed patches (some-file-name# or some-file-name.rej).
85 If there are, either you or I have made a mistake.
86
87 Unlike patches for the 4.x kernels, patches for the 4.x.y kernels
88 (also known as the -stable kernels) are not incremental but instead apply
89 directly to the base 4.x kernel. For example, if your base kernel is 4.0
90 and you want to apply the 4.0.3 patch, you must not first apply the 4.0.1
91 and 4.0.2 patches. Similarly, if you are running kernel version 4.0.2 and
92 want to jump to 4.0.3, you must first reverse the 4.0.2 patch (that is,
93 patch -R) _before_ applying the 4.0.3 patch. You can read more on this in
94 Documentation/applying-patches.txt
95
96 Alternatively, the script patch-kernel can be used to automate this
97 process. It determines the current kernel version and applies any
98 patches found.
99
100 linux/scripts/patch-kernel linux
101
102 The first argument in the command above is the location of the
103 kernel source. Patches are applied from the current directory, but
104 an alternative directory can be specified as the second argument.
105
106 - Make sure you have no stale .o files and dependencies lying around:
107
108 cd linux
109 make mrproper
110
111 You should now have the sources correctly installed.
112
113SOFTWARE REQUIREMENTS
114
115 Compiling and running the 4.x kernels requires up-to-date
116 versions of various software packages. Consult
117 Documentation/Changes for the minimum version numbers required
118 and how to get updates for these packages. Beware that using
119 excessively old versions of these packages can cause indirect
120 errors that are very difficult to track down, so don't assume that
121 you can just update packages when obvious problems arise during
122 build or operation.
123
124BUILD directory for the kernel:
125
126 When compiling the kernel, all output files will per default be
127 stored together with the kernel source code.
128 Using the option "make O=output/dir" allows you to specify an alternate
129 place for the output files (including .config).
130 Example:
131
132 kernel source code: /usr/src/linux-4.X
133 build directory: /home/name/build/kernel
134
135 To configure and build the kernel, use:
136
137 cd /usr/src/linux-4.X
138 make O=/home/name/build/kernel menuconfig
139 make O=/home/name/build/kernel
140 sudo make O=/home/name/build/kernel modules_install install
141
142 Please note: If the 'O=output/dir' option is used, then it must be
143 used for all invocations of make.
144
145CONFIGURING the kernel:
146
147 Do not skip this step even if you are only upgrading one minor
148 version. New configuration options are added in each release, and
149 odd problems will turn up if the configuration files are not set up
150 as expected. If you want to carry your existing configuration to a
151 new version with minimal work, use "make oldconfig", which will
152 only ask you for the answers to new questions.
153
154 - Alternative configuration commands are:
155
156 "make config" Plain text interface.
157
158 "make menuconfig" Text based color menus, radiolists & dialogs.
159
160 "make nconfig" Enhanced text based color menus.
161
162 "make xconfig" Qt based configuration tool.
163
164 "make gconfig" GTK+ based configuration tool.
165
166 "make oldconfig" Default all questions based on the contents of
167 your existing ./.config file and asking about
168 new config symbols.
169
170 "make silentoldconfig"
171 Like above, but avoids cluttering the screen
172 with questions already answered.
173 Additionally updates the dependencies.
174
175 "make olddefconfig"
176 Like above, but sets new symbols to their default
177 values without prompting.
178
179 "make defconfig" Create a ./.config file by using the default
180 symbol values from either arch/$ARCH/defconfig
181 or arch/$ARCH/configs/${PLATFORM}_defconfig,
182 depending on the architecture.
183
184 "make ${PLATFORM}_defconfig"
185 Create a ./.config file by using the default
186 symbol values from
187 arch/$ARCH/configs/${PLATFORM}_defconfig.
188 Use "make help" to get a list of all available
189 platforms of your architecture.
190
191 "make allyesconfig"
192 Create a ./.config file by setting symbol
193 values to 'y' as much as possible.
194
195 "make allmodconfig"
196 Create a ./.config file by setting symbol
197 values to 'm' as much as possible.
198
199 "make allnoconfig" Create a ./.config file by setting symbol
200 values to 'n' as much as possible.
201
202 "make randconfig" Create a ./.config file by setting symbol
203 values to random values.
204
205 "make localmodconfig" Create a config based on current config and
206 loaded modules (lsmod). Disables any module
207 option that is not needed for the loaded modules.
208
209 To create a localmodconfig for another machine,
210 store the lsmod of that machine into a file
211 and pass it in as a LSMOD parameter.
212
213 target$ lsmod > /tmp/mylsmod
214 target$ scp /tmp/mylsmod host:/tmp
215
216 host$ make LSMOD=/tmp/mylsmod localmodconfig
217
218 The above also works when cross compiling.
219
220 "make localyesconfig" Similar to localmodconfig, except it will convert
221 all module options to built in (=y) options.
222
223 You can find more information on using the Linux kernel config tools
224 in Documentation/kbuild/kconfig.txt.
225
226 - NOTES on "make config":
227
228 - Having unnecessary drivers will make the kernel bigger, and can
229 under some circumstances lead to problems: probing for a
230 nonexistent controller card may confuse your other controllers
231
232 - Compiling the kernel with "Processor type" set higher than 386
233 will result in a kernel that does NOT work on a 386. The
234 kernel will detect this on bootup, and give up.
235
236 - A kernel with math-emulation compiled in will still use the
237 coprocessor if one is present: the math emulation will just
238 never get used in that case. The kernel will be slightly larger,
239 but will work on different machines regardless of whether they
240 have a math coprocessor or not.
241
242 - The "kernel hacking" configuration details usually result in a
243 bigger or slower kernel (or both), and can even make the kernel
244 less stable by configuring some routines to actively try to
245 break bad code to find kernel problems (kmalloc()). Thus you
246 should probably answer 'n' to the questions for "development",
247 "experimental", or "debugging" features.
248
249COMPILING the kernel:
250
251 - Make sure you have at least gcc 3.2 available.
252 For more information, refer to Documentation/Changes.
253
254 Please note that you can still run a.out user programs with this kernel.
255
256 - Do a "make" to create a compressed kernel image. It is also
257 possible to do "make install" if you have lilo installed to suit the
258 kernel makefiles, but you may want to check your particular lilo setup first.
259
260 To do the actual install, you have to be root, but none of the normal
261 build should require that. Don't take the name of root in vain.
262
263 - If you configured any of the parts of the kernel as `modules', you
264 will also have to do "make modules_install".
265
266 - Verbose kernel compile/build output:
267
268 Normally, the kernel build system runs in a fairly quiet mode (but not
269 totally silent). However, sometimes you or other kernel developers need
270 to see compile, link, or other commands exactly as they are executed.
271 For this, use "verbose" build mode. This is done by passing
272 "V=1" to the "make" command, e.g.
273
274 make V=1 all
275
276 To have the build system also tell the reason for the rebuild of each
277 target, use "V=2". The default is "V=0".
278
279 - Keep a backup kernel handy in case something goes wrong. This is
280 especially true for the development releases, since each new release
281 contains new code which has not been debugged. Make sure you keep a
282 backup of the modules corresponding to that kernel, as well. If you
283 are installing a new kernel with the same version number as your
284 working kernel, make a backup of your modules directory before you
285 do a "make modules_install".
286
287 Alternatively, before compiling, use the kernel config option
288 "LOCALVERSION" to append a unique suffix to the regular kernel version.
289 LOCALVERSION can be set in the "General Setup" menu.
290
291 - In order to boot your new kernel, you'll need to copy the kernel
292 image (e.g. .../linux/arch/i386/boot/bzImage after compilation)
293 to the place where your regular bootable kernel is found.
294
295 - Booting a kernel directly from a floppy without the assistance of a
296 bootloader such as LILO, is no longer supported.
297
298 If you boot Linux from the hard drive, chances are you use LILO, which
299 uses the kernel image as specified in the file /etc/lilo.conf. The
300 kernel image file is usually /vmlinuz, /boot/vmlinuz, /bzImage or
301 /boot/bzImage. To use the new kernel, save a copy of the old image
302 and copy the new image over the old one. Then, you MUST RERUN LILO
303 to update the loading map! If you don't, you won't be able to boot
304 the new kernel image.
305
306 Reinstalling LILO is usually a matter of running /sbin/lilo.
307 You may wish to edit /etc/lilo.conf to specify an entry for your
308 old kernel image (say, /vmlinux.old) in case the new one does not
309 work. See the LILO docs for more information.
310
311 After reinstalling LILO, you should be all set. Shutdown the system,
312 reboot, and enjoy!
313
314 If you ever need to change the default root device, video mode,
315 ramdisk size, etc. in the kernel image, use the 'rdev' program (or
316 alternatively the LILO boot options when appropriate). No need to
317 recompile the kernel to change these parameters.
318
319 - Reboot with the new kernel and enjoy.
320
321IF SOMETHING GOES WRONG:
322
323 - If you have problems that seem to be due to kernel bugs, please check
324 the file MAINTAINERS to see if there is a particular person associated
325 with the part of the kernel that you are having trouble with. If there
326 isn't anyone listed there, then the second best thing is to mail
327 them to me (torvalds@linux-foundation.org), and possibly to any other
328 relevant mailing-list or to the newsgroup.
329
330 - In all bug-reports, *please* tell what kernel you are talking about,
331 how to duplicate the problem, and what your setup is (use your common
332 sense). If the problem is new, tell me so, and if the problem is
333 old, please try to tell me when you first noticed it.
334
335 - If the bug results in a message like
336
337 unable to handle kernel paging request at address C0000010
338 Oops: 0002
339 EIP: 0010:XXXXXXXX
340 eax: xxxxxxxx ebx: xxxxxxxx ecx: xxxxxxxx edx: xxxxxxxx
341 esi: xxxxxxxx edi: xxxxxxxx ebp: xxxxxxxx
342 ds: xxxx es: xxxx fs: xxxx gs: xxxx
343 Pid: xx, process nr: xx
344 xx xx xx xx xx xx xx xx xx xx
345
346 or similar kernel debugging information on your screen or in your
347 system log, please duplicate it *exactly*. The dump may look
348 incomprehensible to you, but it does contain information that may
349 help debugging the problem. The text above the dump is also
350 important: it tells something about why the kernel dumped code (in
351 the above example, it's due to a bad kernel pointer). More information
352 on making sense of the dump is in Documentation/oops-tracing.txt
353
354 - If you compiled the kernel with CONFIG_KALLSYMS you can send the dump
355 as is, otherwise you will have to use the "ksymoops" program to make
356 sense of the dump (but compiling with CONFIG_KALLSYMS is usually preferred).
357 This utility can be downloaded from
358 ftp://ftp.<country>.kernel.org/pub/linux/utils/kernel/ksymoops/ .
359 Alternatively, you can do the dump lookup by hand:
360
361 - In debugging dumps like the above, it helps enormously if you can
362 look up what the EIP value means. The hex value as such doesn't help
363 me or anybody else very much: it will depend on your particular
364 kernel setup. What you should do is take the hex value from the EIP
365 line (ignore the "0010:"), and look it up in the kernel namelist to
366 see which kernel function contains the offending address.
367
368 To find out the kernel function name, you'll need to find the system
369 binary associated with the kernel that exhibited the symptom. This is
370 the file 'linux/vmlinux'. To extract the namelist and match it against
371 the EIP from the kernel crash, do:
372
373 nm vmlinux | sort | less
374
375 This will give you a list of kernel addresses sorted in ascending
376 order, from which it is simple to find the function that contains the
377 offending address. Note that the address given by the kernel
378 debugging messages will not necessarily match exactly with the
379 function addresses (in fact, that is very unlikely), so you can't
380 just 'grep' the list: the list will, however, give you the starting
381 point of each kernel function, so by looking for the function that
382 has a starting address lower than the one you are searching for but
383 is followed by a function with a higher address you will find the one
384 you want. In fact, it may be a good idea to include a bit of
385 "context" in your problem report, giving a few lines around the
386 interesting one.
387
388 If you for some reason cannot do the above (you have a pre-compiled
389 kernel image or similar), telling me as much about your setup as
390 possible will help. Please read the REPORTING-BUGS document for details.
391
392 - Alternatively, you can use gdb on a running kernel. (read-only; i.e. you
393 cannot change values or set break points.) To do this, first compile the
394 kernel with -g; edit arch/i386/Makefile appropriately, then do a "make
395 clean". You'll also need to enable CONFIG_PROC_FS (via "make config").
396
397 After you've rebooted with the new kernel, do "gdb vmlinux /proc/kcore".
398 You can now use all the usual gdb commands. The command to look up the
399 point where your system crashed is "l *0xXXXXXXXX". (Replace the XXXes
400 with the EIP value.)
401
402 gdb'ing a non-running kernel currently fails because gdb (wrongly)
403 disregards the starting offset for which the kernel is compiled.
404